Multimodal Protein Language Models for Enzyme Kinetic Parameters: From Substrate Recognition to Conformational Adaptation

• URL: http://arxiv.org/abs/2603.12845v1
• Date: Fri, 13 Mar 2026 09:45:41 GMT
• Title: Multimodal Protein Language Models for Enzyme Kinetic Parameters: From Substrate Recognition to Conformational Adaptation
• Abstract summary: Predicting enzyme kinetic parameters quantifies how efficiently an enzyme catalyzes a specific substrate.<n>Many learning pipelines simplify this process to a static compatibility problem between the enzyme and substrate.<n>We introduce the Enzyme-Reaction Bridging Adapter (ERBA), which injects cross-modal information via fine-tuning into Protein Language Models.
• Score: 12.729882993671106
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Predicting enzyme kinetic parameters quantifies how efficiently an enzyme catalyzes a specific substrate under defined biochemical conditions. Canonical parameters such as the turnover number ($k_\text{cat}$), Michaelis constant ($K_\text{m}$), and inhibition constant ($K_\text{i}$) depend jointly on the enzyme sequence, the substrate chemistry, and the conformational adaptation of the active site during binding. Many learning pipelines simplify this process to a static compatibility problem between the enzyme and substrate, fusing their representations through shallow operations and regressing a single value. Such formulations overlook the staged nature of catalysis, which involves both substrate recognition and conformational adaptation. In this regard, we reformulate kinetic prediction as a staged multimodal conditional modeling problem and introduce the Enzyme-Reaction Bridging Adapter (ERBA), which injects cross-modal information via fine-tuning into Protein Language Models (PLMs) while preserving their biochemical priors. ERBA performs conditioning in two stages: Molecular Recognition Cross-Attention (MRCA) first injects substrate information into the enzyme representation to capture specificity; Geometry-aware Mixture-of-Experts (G-MoE) then integrates active-site structure and routes samples to pocket-specialized experts to reflect induced fit. To maintain semantic fidelity, Enzyme-Substrate Distribution Alignment (ESDA) enforces distributional consistency within the PLM manifold in a reproducing kernel Hilbert space. Experiments across three kinetic endpoints and multiple PLM backbones, ERBA delivers consistent gains and stronger out-of-distribution performance compared with sequence-only and shallow-fusion baselines, offering a biologically grounded route to scalable kinetic prediction and a foundation for adding cofactors, mutations, and time-resolved structural cues.

Related papers

• Retrieval-Augmented Multimodal Learning for Enzyme-Substrate Interaction Prediction Under Low-Homology Shift [15.372448178786797]
  RAMMESI is a retrieval-augmented multimodal framework for robust prediction.<n>It learns explicit pairwise enzyme-substrate representations through directional cross-modal interaction modeling.<n>It retrieves neighboring enzymes at inference time, recombines them with the substrate, and aggregates the resulting pairwise predictions as contextual evidence.
  arXiv  Detail & Related papers  (2026-06-22T04:01:18Z)
• scHelix: Asymmetric Dual-Stream Integration via Explicit Gene-Level Disentanglement [84.88065404629079]
  scHelix is a dataset-adaptive framework that fundamentally changes how features are processed.<n> scHelix utilizes a dual-stream sparse diffusion encoder equipped with stop-gradient graph caching.<n> scHelix outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-05-18T15:55:13Z)
• RXNRECer Enables Fine-grained Enzymatic Function Annotation through Active Learning and Protein Language Models [20.176405299884824]
  RXNRECer is a transformer-based ensemble framework that directly predicts enzyme-catalyzed reactions without relying on EC numbers.<n>It integrates protein language modeling and active learning to capture both high-level sequence semantics and fine-grained transformation patterns.
  arXiv  Detail & Related papers  (2026-03-13T06:20:14Z)
• Mechanisms of AI Protein Folding in ESMFold [60.598995350547]
  We trace how ESMFold folds a beta hairpin, a prevalent structural motif.<n>We demonstrate that the mechanisms underlying structural decisions of ESMFold can be localized, traced through interpretable representations, and manipulated with strong causal effects.
  arXiv  Detail & Related papers  (2026-02-05T18:54:54Z)
• Pseudodata-guided Invariant Representation Learning Boosts the Out-of-Distribution Generalization in Enzymatic Kinetic Parameter Prediction [12.238915133864046]
  O$2$DENet is a lightweight, plug-and-play module that enhances OOD generalization.<n>O$2$DENet introduces enzyme-substrate perturbations and enforces consistency between original and augmented enzyme-substrate-pair representations.
  arXiv  Detail & Related papers  (2026-01-12T07:03:07Z)
• EnzyCLIP: A Cross-Attention Dual Encoder Framework with Contrastive Learning for Predicting Enzyme Kinetic Constants [2.92594095183629]
  We present EnzyCLIP, a novel dual-encoder framework to predict enzyme kinetic parameters from protein sequences and substrate molecular structures.<n>The model is trained on the CatPred-DB database containing 23,151 Kcat and 41,174 Km experimentally validated measurements.<n>XGBoost ensemble methods applied to the learned embeddings further improved Km prediction (R2 = 0.61) while maintaining robust Kcat performance.
  arXiv  Detail & Related papers  (2025-11-29T08:13:06Z)
• Multimodal Regression for Enzyme Turnover Rates Prediction [57.60697333734054]
  We propose a framework for predicting the enzyme turnover rate by integrating enzyme sequences, substrate structures, and environmental factors.<n>Our model combines a pre-trained language model and a convolutional neural network to extract features from protein sequences.<n>We leverage symbolic regression via Kolmogorov-Arnold Networks to explicitly learn mathematical formulas that govern the enzyme turnover rate.
  arXiv  Detail & Related papers  (2025-09-15T11:07:26Z)
• OmniESI: A unified framework for enzyme-substrate interaction prediction with progressive conditional deep learning [46.402707495664174]
  We introduce a two-stage progressive framework, OmniESI, for enzyme-substrate interaction prediction through conditional deep learning.<n>We show that OmniESI consistently delivered superior performance than state-of-the-art specialized methods.<n>Overall, OmniESI represents a unified predictive approach for enzyme-substrate interactions.
  arXiv  Detail & Related papers  (2025-06-22T09:40:40Z)
• Reaction-conditioned De Novo Enzyme Design with GENzyme [64.14088142258498]
  textscGENzyme is a textitde novo enzyme design model that takes a catalytic reaction as input and generates the catalytic pocket, full enzyme structure, and enzyme-substrate binding complex.<n>textscGENzyme is an end-to-end, three-staged model that integrates (1) a catalytic pocket generation and sequence co-design module, (2) a pocket inpainting and enzyme inverse folding module, and (3) a binding and screening module to optimize and predict enzyme-substrate complexes.
  arXiv  Detail & Related papers  (2024-11-10T00:37:26Z)
• Structure Language Models for Protein Conformation Generation [66.42864253026053]
  Traditional physics-based simulation methods often struggle with sampling equilibrium conformations.<n>Deep generative models have shown promise in generating protein conformations as a more efficient alternative.<n>We introduce Structure Language Modeling as a novel framework for efficient protein conformation generation.
  arXiv  Detail & Related papers  (2024-10-24T03:38:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.